Key facts
- The Authors Guild advises against using generative AI to mimic other writers' unique styles, voices, or attributes.
- Mimicking an author's style may lead to legal issues like unfair competition or copyright infringement claims.
- OpenAI's DALL-E 3 model is designed to refuse requests for images in the style of living artists.
- A study found Google's Gemini consistently complied with requests to copy author styles, while Perplexity AI refused.
- ChatGPT users report the model now declines prompts to generate content in specific authors' styles.
ChatGPT has reportedly ceased cloning the distinct voices and styles of famous writers, aligning with ethical guidelines that discourage profiting from or harming the value of another author's work through mimicry. The Authors Guild has published best practices urging writers to respect their peers and avoid using generative AI to copy unique attributes, warning that such actions could lead to claims of unfair competition or copyright infringement.
This move comes after several public instances where AI's stylistic mimicry drew criticism. Last year, author Lena McDonald faced backlash when her book contained an AI-generated passage that explicitly stated it was rewritten to align with another author's style. While OpenAI's image generation model, DALL-E 3, is designed to decline requests for images in the style of living artists, a December model specification did not explicitly prohibit replicating copyrighted material or mimicking style in written responses.
However, not all large language models (LLMs) handle these requests uniformly. A study by No Latency found that Google's Gemini consistently complied with requests to copy author styles. In contrast, Perplexity AI's models consistently refused such requests, redirecting users similarly to how ChatGPT now operates. Anthropic's Claude and Microsoft's Copilot reportedly complied with style-mimicry requests but included qualifications, indicating an awareness of the imitation issue.
For some users, these changes by OpenAI have created difficulties. One Reddit user, Dazzling-Major-5620, expressed frustration, stating that ChatGPT now refuses to generate content in specific authors' styles, which had been a valuable feature for their prompts. The user is exploring workarounds, such as feeding the AI their own edited prose.